Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
11-13-2018, 01:22 PM   #16
Site Supporter
Site Supporter
gofour3's Avatar

Join Date: Oct 2008
Location: Vancouver, Canada
Posts: 6,248
QuoteOriginally posted by BruceBanner Quote
So with the 28/2.8, does that mean when choosing F8, and lining the orange markings up (10ft/3m), that any subject standing 5ft/1.5m (half the distance) from the camera/lens would be in focus (and then out to infinity)?
Correct.

Phil.

11-13-2018, 04:05 PM   #17
dms
Site Supporter
Site Supporter




Join Date: Aug 2011
Location: New York, NY
Photos: Gallery
Posts: 1,880
QuoteOriginally posted by BruceBanner Quote
So I'm still a little confused (surprise surprise) over the blurb;

"Focus will be sharp for all subjects located from about half of the red marked distance all the way out to infinity without any adjustment."

So with the 28/2.8, does that mean when choosing F8, and lining the orange markings up (10ft/3m), that any subject standing 5ft/1.5m (half the distance) from the camera/lens would be in focus (and then out to infinity)?
Perhaps to add one thing to what Phil said,
-- yes if using FF (35mm film)
--if APSC it is for 1 stop closed more--i.e., f/11.

---------- Post added 11-13-18 at 04:12 PM ----------

QuoteOriginally posted by cpk Quote
I would question your conclusions about modern gear taking us further, but you are right about image size and viewing distance. Apparent sharpness depends entirely on the resolving power of the human eye + viewing distance. One would expect to stand further away from a 16 x 20 print than from an 8 x 10 one. If you view the former at the same distance as you would normally view the latter, apparent depth of field lessens in the larger print. Pixel peeping is meaningless in this context unless you stand away from the screen at a distance that adjusts for the image size on the screen when compared to the print size.
I think this is an important point.

I would add if you find you print large and want to look closer, then use the values of DOF and H for 1 stop more open than actually lens aperture. So in our example use f/16 and the DOF and H values for f/8.

If using different sensor size we do similar. We can adjust f-stop rather than recalculating DOF and H.

So if you just discard the concept you are missing a useful procedure, Better to adjust it.
11-13-2018, 06:05 PM - 1 Like   #18
Site Supporter
Site Supporter




Join Date: Nov 2015
Photos: Albums
Posts: 1,372
QuoteOriginally posted by cpk Quote
I would question your conclusions about modern gear taking us further, but you are right about image size and viewing distance. Apparent sharpness depends entirely on the resolving power of the human eye + viewing distance. One would expect to stand further away from a 16 x 20 print than from an 8 x 10 one. If you view the former at the same distance as you would normally view the latter, apparent depth of field lessens in the larger print. Pixel peeping is meaningless in this context unless you stand away from the screen at a distance that adjusts for the image size on the screen when compared to the print size.
Sorry, I wrote the comment while using the phone and was perhaps a bit too concise... maybe this is more complete...

If I grab my 28mm SMC-A lens and my K1000 and shoot some Ektar 100 film, and make 8*10 inch prints with the negs, holding them at approximately arms-length to view them, the amount in focus will agree approximately with the depth of field scale on the lens and the orange hyperfocal point.

That's my understanding of the assumptions that went in to most depth of field scales (though I've heard of different ones...)

If I take the same lens on my K1 and make some 8*10 inch prints of the same shot, viewing them the same way, I'll get basically the same result when it comes to focus.

If I sit down at the computer and look at a scan of my Ektar 100 that my local lab gives me (6Mp), and zoom all the way in (1:1), the area of focus on my monitor will be about the same as it was on the 8*10 inch print because I've exceeded the resolution of the scan with the zoom. In this case, I'll again see in focus about what the lens suggests, if not a bit more. I can't see anything beyond that because I lack the image resolution.

If I then sit down with the 36Mp image from the K1 and zoom all the way in, there are going to be areas that are out of focus on the screen but within the range of the depth of field scale. In this case I have vastly more resolution than was expected when the scales were done decades ago, and by zooming in on the monitor, I'm basically looking at a portion of a virtual enlargement from closer than I should.

Basically, what I've done with the K1 is blow the assumptions that the depth of field scale was created using.

I've noticed that the K1 is good (or bad, I guess, depending) at blowing a lot of film-era assumptions...

Of course, that means it also makes some really good 8*10 inch prints

-Eric
11-13-2018, 08:17 PM - 1 Like   #19
cpk
Site Supporter
Site Supporter
cpk's Avatar

Join Date: Feb 2010
Location: Montreal
Posts: 366
QuoteOriginally posted by TwoUptons Quote
If I then sit down with the 36Mp image from the K1 and zoom all the way in, there are going to be areas that are out of focus on the screen but within the range of the depth of field scale. In this case I have vastly more resolution than was expected when the scales were done decades ago, and by zooming in on the monitor, I'm basically looking at a portion of a virtual enlargement from closer than I should.

Basically, what I've done with the K1 is blow the assumptions that the depth of field scale was created using.

I've noticed that the K1 is good (or bad, I guess, depending) at blowing a lot of film-era assumptions...
I still cannot agree. The concept of depth of field depends on the ability of the eye to see fine detail at varying distances. It is entirely an optical concept and has nothing to do with the resolution of the sensor. Whether we are talking film or digital, the apparent sharpness depends entirely on the size of the circle of confusion at varying distances within the image. At the focus plane the circle of confusion ideally is a point (in a perfect optical system) and gets larger as one moves away from the focus plane, both in front of it and behind it. Closer to the focus plane the eye sees these circles of confusion as points of sharpness; but as parts of the image move away from the focal plane at some point the circles of confusion become too large for the eye to see as sharp. This point (two actually), both in front of the focal plane and behind, marks the limit of the depth of field for that lens setting.

QuoteOriginally posted by dms Quote
I would add if you find you print large and want to look closer, then use the values of DOF and H for 1 stop more open than actually lens aperture. So in our example use f/16 and the DOF and H values for f/8.
This suggestion is effectively saying to use a smaller circle of confusion in calculating the depth of field.

Here is the explanation from Wikipedia (I looked this up after writing my explanation in case I wasn't clear enough):

Precise focus is possible in only one two-dimensional plane; in that plane, a point object will produce a point image.[1] In any other plane, a point object is defocused, and will produce a blur spot shaped like the aperture, which for the purpose of analysis is usually assumed to be circular. When this circular spot is sufficiently small, it is indistinguishable from a point, and appears to be in focus; it is rendered as "acceptably sharp". The diameter of the circle increases with distance from the plane of focus; the largest circle that is indistinguishable from a point is known as the acceptable circle of confusion, or informally, simply as the circle of confusion. The acceptable circle of confusion is influenced by visual acuity, viewing conditions, and the amount by which the image is enlarged (Ray 2000, 52–53). The increase of the circle diameter with defocus is gradual, so the limits of depth of field are not hard boundaries between sharp and unsharp.


Last edited by cpk; 11-15-2018 at 04:35 PM. Reason: spelling error corrected; extraneous word removed
11-13-2018, 10:44 PM   #20
Otis Memorial Pentaxian
Loyal Site Supporter
stevebrot's Avatar

Join Date: Mar 2007
Location: Vancouver (USA)
Photos: Gallery | Albums
Posts: 33,944
QuoteOriginally posted by cpk Quote
I still cannot agree. The concept of depth of field depends on the ability of the eye to see fine detail at varying distances.
You are correct.

DOF is calculated based on several constants and and a few variables. System resolution is not one of the variables and neither is format size per se.


Steve
11-15-2018, 04:41 PM   #21
cpk
Site Supporter
Site Supporter
cpk's Avatar

Join Date: Feb 2010
Location: Montreal
Posts: 366
Thanks, Steve, for the confirmation; I haven't provided this type of explanation for many years. I should add that this explanation got me a career in IT which exceeded 50 years.
11-15-2018, 06:43 PM   #22
Otis Memorial Pentaxian
Loyal Site Supporter
stevebrot's Avatar

Join Date: Mar 2007
Location: Vancouver (USA)
Photos: Gallery | Albums
Posts: 33,944
QuoteOriginally posted by cpk Quote
I should add that this explanation got me a career in IT which exceeded 50 years.
50 years...wow! The changes you witnessed must have been incredible. I am old enough to have used key-punch for FORTRAN programming, but did not see a REAL computer (PDP-11) before 1973 when a Freshman in college.


Steve
11-17-2018, 08:35 AM   #23
Site Supporter
Site Supporter




Join Date: Nov 2015
Photos: Albums
Posts: 1,372
Part 1...

So in an attempt to suggest that yes, the resolution of the sensor can have something to do with hyperfocal performance... here's part 1...
Part 2 is currently residing in my KX, and will be posted when I finish the roll, probably Tuesday or so (I hope)...

So, Pentax SMC-A 28mm f2.8 at f8 shot at the hyperfocal distance on a K1 this morning... (photo sized by Apple ahead of posting)... with a hood off a Takumar 28mm f3.5 and no filters



And a crop of the flower... it's at 3m, the grass in the neighbor's yard is about 8m, the rubbish bins are about 25m, and the basketball goal is about 41m



Now set at infinity focus...





Just from these, you can tell that the markings on the lens are not sufficient to cover the K1s resolution (and that's even with SOOC JPEGs at ISO 400, so not its best)
f11 may have worked better, but I didn't do that one... and I'm not sure if I may have had other softness/diffraction issues with the smaller aperture...

Interestingly, when the lens is at infinity, the flower is almost there, but the grass (and certainly the bins) look vastly better. I wasn't expecting the grass (at ~8m) to be so sensitive.
And the lens markings put the f8 depth-of-field boundary at the same orange 10ft/3m mark when the lens is at infinity, so in theory the flower should be about in focus...

Now, yes, I have absolutely and quite intentionally pixel-peeped these. If printed at 10*15cm, I doubt you could really tell the difference at an appropriate viewing distance.
But I think I have shown that at least for my sample of the lens, the hyperfocal markings are not sufficient on a high-resolution body if you pixel-peep or print large...

And, of course, the solution is simple... either use a smaller aperture (and I'm guessing one stop is probably enough here) and the same DoF range or use the same aperture and a smaller DoF range.
I usually use the latter, assuming, for example the f5.6 lines will work for f8 and focusing accordingly. That seems to work OK, but I don't do it often...
For reference, the SMC-A 28mm f2.8 lines put that at about 2.5m and out with the infinity mark on f5.6 as opposed to 1.4m and out with the f8 marks.

The second part of my theory, which I can't prove until my film is done, is that this is really partially resolution-dependent... and that the lower resolution of the film will 'blur the lines' of the hyperfocal limits...

For that, we'll see what Fuji and Noritsu have to contribute in a few days...

And in the world of weird, the only 35mm Pentax with film in it this morning was the KX, with the 28mm already mounted... as if it knew ahead of time...

-Eric

11-17-2018, 10:59 AM   #24
dms
Site Supporter
Site Supporter




Join Date: Aug 2011
Location: New York, NY
Photos: Gallery
Posts: 1,880
QuoteOriginally posted by TwoUptons Quote
So in an attempt to suggest that yes, the resolution of the sensor can have something to do with hyperfocal performance... here's part 1...
Part 2 is currently residing in my KX, and will be posted when I finish the roll, probably Tuesday or so (I hope)...

So, Pentax SMC-A 28mm f2.8 at f8 shot at the hyperfocal distance on a K1 this morning... (photo sized by Apple ahead of posting)... with a hood off a Takumar 28mm f3.5 and no filters



And a crop of the flower... it's at 3m, the grass in the neighbor's yard is about 8m, the rubbish bins are about 25m, and the basketball goal is about 41m



Now set at infinity focus...





Just from these, you can tell that the markings on the lens are not sufficient to cover the K1s resolution (and that's even with SOOC JPEGs at ISO 400, so not its best)
f11 may have worked better, but I didn't do that one... and I'm not sure if I may have had other softness/diffraction issues with the smaller aperture...

Interestingly, when the lens is at infinity, the flower is almost there, but the grass (and certainly the bins) look vastly better. I wasn't expecting the grass (at ~8m) to be so sensitive.
And the lens markings put the f8 depth-of-field boundary at the same orange 10ft/3m mark when the lens is at infinity, so in theory the flower should be about in focus...

Now, yes, I have absolutely and quite intentionally pixel-peeped these. If printed at 10*15cm, I doubt you could really tell the difference at an appropriate viewing distance.
But I think I have shown that at least for my sample of the lens, the hyperfocal markings are not sufficient on a high-resolution body if you pixel-peep or print large...

And, of course, the solution is simple... either use a smaller aperture (and I'm guessing one stop is probably enough here) and the same DoF range or use the same aperture and a smaller DoF range.
I usually use the latter, assuming, for example the f5.6 lines will work for f8 and focusing accordingly. That seems to work OK, but I don't do it often...
For reference, the SMC-A 28mm f2.8 lines put that at about 2.5m and out with the infinity mark on f5.6 as opposed to 1.4m and out with the f8 marks.

The second part of my theory, which I can't prove until my film is done, is that this is really partially resolution-dependent... and that the lower resolution of the film will 'blur the lines' of the hyperfocal limits...

For that, we'll see what Fuji and Noritsu have to contribute in a few days...

And in the world of weird, the only 35mm Pentax with film in it this morning was the KX, with the 28mm already mounted... as if it knew ahead of time...

-Eric
Two things:
-- Firstly if the resolution is not the limiting factor the greatest detail is at the particular distance set--so if (a) focused/set for infinity versus (b) something closer, pixel peeping will show the closer areas to be sharper with (b) and infinity with (a). Pixel peeping generally puts you outside any hyperfocal distance or DOF attributes, as it is (usually) a far larger enlargement than about 8x. But if viewed/printed at about 8x enlargement they will both look acceptably sharp in the areas where the DOF overlaps; however, setting at infinity gives you half the DOF as setting at the hyperfocal distance.
-- Secondly, if the resolution is low enough (either due to sensor/film or lens) it will appear unsharp no matter the DOF. A pinhole lens has almost all distances in its "DOF" (although DOF is the wrong term) and yet everything is unsharp.

BTW if you want to pixel peep and you decide the enlargement to be used, you can still use the hyperfocal distance scale on your (FF) lens, just use a different value consistent with the enlargement. If you plan to pixel peep at 16X enlargement then use the hyperfocal distance for two stops more open than actually setting lens at. We (I anyway) already do that when using a FF lens on a 1.5 cropped sensor, as we use the hyperfocal distance at 1 stop more open than actually setting lens at. The number of stops difference is about:

(actual enlargement/8) stops

Last edited by dms; 11-17-2018 at 11:16 AM. Reason: Added BTW ...., also clarify DOF for pinhole lens
11-17-2018, 05:42 PM   #25
Site Supporter
Site Supporter
BruceBanner's Avatar

Join Date: Dec 2015
Location: Blue Mountains, NSW, Australia
Posts: 3,930
Original Poster
I did some testing today as well. I used my old and tattered mailbox as my 'pretend person', I roughly centred the composition as I didn't want edge sharpness to interfere (at this point in my experimentation).

I started off by getting my measuring tape and roughly finding 2.2m back from the face of the mail box (you can see the tape measure line on the ground ).

I then tripodded up and started at f2.8 and using LV got focus on the face of the mail box, indeed it was very close to the orange 7ft/2.2m markings on the lens. I used Av mode throughout with timer 2 sec mode.

F2.8, 7ft/2.2m

Decent focus obtained, I never find this lens particularly sharp, pixelshifting can really help this lens shine however.

Moving on...

Next up same focus again, F8, 7ft/2.2m


I was quite surprised to actually see things are still quite soft past the mail box, even the grass and boulder past the step is soft, the DoF is still quite narrow here. I'm not sure I would feel confident at using these orange markings guidelines to getting a snapshot if this is the result, I still clearly need to be quite specific to my distance to subject, should the mail box move a tad back, or the subject be 5m away... pretty soft image.

Next up same focus again (still have not touched the focus ring since the initial f2.8 shot) but now F11, 7ft/2.2m;


Ok... now things appear quite different, things look quite in focus across the mail box right out! Neat.

Next up, I stay with the camera in the tripod but I now throw the focus ring to infinity and dial back to F8;

Wow... ok, this is starting to look a lot like the F11 with orange markings shot above!, but we have gained valuable stops of light! We've lost perhaps a tad of sharpness on the mail box, but it's still pretty acceptable.

Next up, infinity again and why not... F13;



So... draw what conclusions you like, for me I think just throwing the focus to infinity, using F8 and ensuring that 2.2m clearing distance is not a bad idea. I feel lining up the orange markings (f8+7ft/2.2) is still asking for trouble, you need to be quite accurate with your distance to subject, a metre out or so and I think the shot will not be as well focused as throwing the focus ring into infinity.

Thoughts?
11-17-2018, 07:14 PM   #26
Otis Memorial Pentaxian
Loyal Site Supporter
stevebrot's Avatar

Join Date: Mar 2007
Location: Vancouver (USA)
Photos: Gallery | Albums
Posts: 33,944
QuoteOriginally posted by BruceBanner Quote
Thoughts?
Is it safe to assume that the above examples are not crops?

Trees at center and roof across street appear acceptably sharp in the f/8 @ 2.2m example, though the examples overall appear somewhat less crisp than other photos I have seen taken with the Pentax-A 24/2.8.

I suspect that the infinity calibration of the lens is to a point somewhat closer than one might think, but at 24mm, anything past about 50m looks about the same if one is not pixel peeping or viewing at much over 8x10" enlargement.


Steve
11-19-2018, 10:34 PM   #27
Site Supporter
Site Supporter
BruceBanner's Avatar

Join Date: Dec 2015
Location: Blue Mountains, NSW, Australia
Posts: 3,930
Original Poster
QuoteOriginally posted by stevebrot Quote
Is it safe to assume that the above examples are not crops?

Trees at center and roof across street appear acceptably sharp in the f/8 @ 2.2m example, though the examples overall appear somewhat less crisp than other photos I have seen taken with the Pentax-A 24/2.8.

I suspect that the infinity calibration of the lens is to a point somewhat closer than one might think, but at 24mm, anything past about 50m looks about the same if one is not pixel peeping or viewing at much over 8x10" enlargement.


Steve
Yep, uncropped. I guess I have to work with what I have, f8+Infinity seems like a good compromise if I'm struggling to guess their distance. I need a laser pointer
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
aperture, distance, dof, f8, focus, half, image, infinity, k-mount, lens, location, pentax lens, people, print, screen, shot, shots, size, slr lens, surprise, values, view
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Penta k3 issue writing to card or card issue? Andrewlang Troubleshooting and Beginner Help 19 10-29-2018 06:27 PM
Machinery Orange car, orange sunset AlexBW Post Your Photos! 7 11-14-2017 12:52 PM
Abstract Writing's on the Wall Kerrowdown Post Your Photos! 3 08-13-2017 09:56 AM
K3ii writing black frames on SD card Lolamorena Troubleshooting and Beginner Help 14 04-25-2017 06:21 PM
Difference in writing time between shooting with DA lenses and K lenses pirivirus Pentax K-5 6 04-08-2013 06:18 PM



All times are GMT -7. The time now is 05:40 PM. | See also: NikonForums.com, CanonForums.com part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top